Dynamic Range of Mass Accuracy in LTQ Orbitrap Hybrid Mass Spectrometer
Total Page:16
File Type:pdf, Size:1020Kb
Dynamic Range of Mass Accuracy in LTQ Orbitrap Hybrid Mass Spectrometer Alexander Makarov, Eduard Denisov, Oliver Lange, and Stevan Horning Thermo Electron (Bremen) GmbH, Bremen, Germany Using a novel orbitrap mass spectrometer, the authors investigate the dynamic range over which accurate masses can be determined (extent of mass accuracy) for short duration experiments typical for LC/MS. A linear ion trap is used to selectively fill an intermediate ion storage device (C-trap) with ions of interest, following which the ensemble of ions is injected into an orbitrap mass analyzer and analyzed using image current detection and fast Fourier transformation. Using this technique, it is possible to generate ion populations with intraspec- trum intensity ranges up to 104. All measurements (including ion accumulation and image current detection) were performed in less than1sataresolving power of 30,000. It was shown that 5-ppm mass accuracy of the orbitrap mass analyzer is reached with Ͼ95% probability at a dynamic range of more than 5000, which is at least an order of magnitude higher than typical values for time-of-flight instruments. Due to the high resolving power of the orbitrap, accurate mass of an ion could be determined when the signal was reliably distinguished from noise Ͼ ѧ (S/Np-p 2 3). (J Am Soc Mass Spectrom 2006, 17, 977–982) © 2006 American Society for Mass Spectrometry he dynamic range over which accurate measure- troiding introduced by the noise of the image current ments of mass can be made (“extent of mass preamplifier[5–8].UnlikeTOFs,FTICRemploysmuch Taccuracy”) is a key analytical figure-of-merit for slower acquisition systems with much higher dynamic any accurate-mass analyzer. In practice, such analyzers range. At high intensities, Coulomb repulsion, rather are coupled to liquid chromatography or other separa- than detector saturation, produces mass shifts that tion methods, and measurements are made for transient dependnotonlyonthetotalcharge[9–11]butalsoon signals (e.g., with spectra recorded at a rate of 1 theintensitiesofindividualmasspeaks[12,13].Over- spectrum/s). For any analyzer, mass accuracy is limited all, intrascan dynamic range of a few thousand is statistically by too few ions detected or by peak position possible[8,14]withmassaccuracyofafewppm. shifts due to too many ions. Though being a universal This work investigates the extent of mass accuracy problem, limitations to the extent of mass accuracy have for a novel Fourier transform mass spectrometer: been investigated in detail for time-of-flight (TOF) mass LTQ Orbitrap. This instrument combines a linear ion analyzers. These analyzers are particularly susceptible trap with radial ejection [15] and an orbitrap mass to variations in ion intensity because they use fast analyzer[16].Theorbitrapmassanalyzerisanelec- acquisition systems with inherently modest dynamic trostatic trap wherein tangentially injected ions rotate range [1–4]. Even with lock-mass and analyte signals around a central electrode, being confined by apply- far from saturation, a strong dependence of peak posi- ing an appropriate voltage between the outer and tion on peak intensity is observed. As a result, 5 ppm central electrodes. Mass analysis is based on image r.m.s. mass accuracy cannot be achieved over a signal current detection of frequencies of axial oscillations. range larger than a few hundred in 1 s acquisition in ion Therefore, its extent of mass accuracy is limited by counting TOFs, even when advanced algorithms for the same factors as FT ICR. The objective of this work intensity correction are employed [1–3]. TOFs with is to determine upper and lower limits of ion inten- analog detection are in principle capable of a dynamic sities for accurate mass analysis and ways for their rangeofaroundonethousand[4]. improvement. In Fourier transform ion cyclotron resonance (FT ICR) mass spectrometry, mass accuracy at low signal intensities is limited by the imprecision of peak cen- Experimental All experiments were carried out using a mixture of Published online June 5, 2006 Ultramark 1600 (Lancaster Synthesis Inc., Windham, Address reprint requests to Dr. A. A. Makarov, Thermo Electron (Bremen) GmbH, Hanna Kunath Strasse 11, Bremen 28199, Germany. E-mail: NH) and MRFA peptide in 50:50 vol/vol water/aceto- [email protected] nitrile solution. © 2006 American Society for Mass Spectrometry. Published by Elsevier Inc. Received January 13, 2006 1044-0305/06/$32.00 Revised March 16, 2006 doi:10.1016/j.jasms.2006.03.006 Accepted March 16, 2006 978 MAKAROV ET AL. J Am Soc Mass Spectrom 2006, 17, 977–982 Figure 1. Experimental sequence for measurements of the extent of mass accuracy the orbitrap mass analyzer: (a) Injection of the first set of ions and trapping in the C-trap; (b) injection of the second set of ions and trapping in the C-trap; (c) pulsed injection of mixed ion population into the orbitrap; (d) ion detection in the orbitrap. Results and Discussion and an orbitrap mass analyzer. Key to operation of this system is a C-shaped storage trap, which is used to Instrument Operation store and collisionally cool ions before injection into the ThemassspectrometerdepictedinFigure1isahybrid orbitrap. With this device, ions are pulsed into the system combining a linear ion trap mass spectrometer central point of the C-trap arc that coincides with the J Am Soc Mass Spectrom 2006, 17, 977–982 DYNAMIC RANGE OF LTQ ORBITRAP 979 DR_524low_1e6L_int_profile # 24 RT: 0.39 AV: 1 NL: 1.49E10 T: FTMS + p ESI Full ms2 [email protected] [ 400.00-2000.00] 1421.98035 R=19245 100 90 0.06 0.05 80 0.04 Relative Abundance Relative 70 0.03 524.26611 R=30209 60 0.02 0.01 50 0.00 521 522 523 524 525 526 527 40 m/z Relative Abundance Relative 30 20 10 0 400 600 800 1000 1200 1400 1600 1800 2000 m/z Figure 2. A typical mass spectrum acquired at the ratio 5000:1 of maximum to minimum intensities (external calibration). orbitrap entrance aperture. Ions are captured in the ion populations including, if necessary, internal cali- orbitrap by rapidly increasing the electric field and brants. Mass calibration coefficients were determined detection of image current from coherent ion packets for different AGC target values and interpolated for takes place after voltages have stabilized [16]. Signals intermediate values. No intensity-dependant correc- from each of the orbitrap outer electrodes are amplified tions of m/z were made for data processing. by a differential amplifier and transformed into a fre- The resolving power was reduced to nominal 30,000 quency spectrum by fast Fourier transformation. The (at m/z 400 Th after zero-filling and Kaiser-Bessel apo- frequency spectrum is converted into a mass spectrum dization, 0.38 s transient duration) so that the experi- using a two-point calibration and processed with Xcali- ment cycle time of 1 s still allowed more than sufficient bur software. time to store up to a million of ions in the C-trap. All data below correspond to a single spectrum acquisition. Measurement Methodology To model the widest possible range of conditions, intensities of dominant and minor peaks were varied To explore the extent of mass accuracy of the orbitrap over orders of magnitude to achieve variations of ratio analyzer, it is important to provide a reproducible and of intensities between 1 and 10,000. For internal calibra- as wide as possible spread of signal intensities within tion evaluation, the intense peak was used as the the same spectrum. An effort was made to achieve a calibrant. Here and below, all resolving powers are wide range of signal intensities by using an electrospray presented as full-width half-maximum (FWHM) values. source with widely different concentrations of analytes. However, it appeared that competition between ana- Results of Measurements lytes for protons precluded the electrospray source from producing reliable and controllable signals at Figure2showsatypicalmassspectrumusedtodeter- levels in the range 1:2000 to 1:5000, compared with the mine mass errors at the extreme limits of dynamic major component of the analyte mixture. To investigate range. Target values were 106 for the major peak and properties of the orbitrap analyzer rather than the 102 for the minor peak; however, space charge repulsion electrospray source, another approach was applied. in the C-trap resulted in a significant reduction of the This approach capitalized on the ability of the C-trap to major peak intensity (about 2-fold). The minor peak has store multiple fills from the linear ion trap per injection such a low S/N that noise starts to limit the precision of intotheorbitrap,asillustratedinFigure1.Thenumber mass measurement in agreement with the published of ions in each fill is individually controlled over several literature [5–7]. It this paper, noise is characterized as orders of magnitude using automatic gain control the maximum peak-to-peak amplitude of thermal noise (AGC), while the selection of masses in each fill is of the preamplifier over the full mass range (for exam- regulated using isolation in the linear ion trap. This ple,inFigure2thenoisestaysat0.008%ofthemajor creates a flexible and versatile tool for forming desired peak). For a signal with a low S/N ratio, it was found 980 MAKAROV ET AL. J Am Soc Mass Spectrom 2006, 17, 977–982 External calibration Internal calibration S/N≈2 S/N≈2 DR=5,000 DR=5,000 10.0 10.0 a) 9.0 b) 9.0 8.0 m/z 524.2650 8.0 m/z 524.2650 7.0 7.0 6.0 6.0 5.0 5.0 error, ppm 4.0 4.0 Mass error, ppm Mass 3.0 3.0 2.0 2.0 1.0 1.0 0.0 0.0 1 10 100 1000 10000 1 10 100 1000 10000 Intensity ratio Intensity ratio c) 10.0 10.0 9.0 d) 9.0 8.0 m/z 1121.9970 8.0 m/z 1121.9970 7.0 7.0 m 6.0 6.0 5.0 5.0 rror, ppm error, pp 4.0 4.0 Mass e 3.0 Mass 3.0 2.0 2.0 1.0 1.0 0.0 0.0 1 10 100 1000 10000 1 10 100 1000 10000 IntensityMax/min intensity ratio Max/minIntensity intensity ratio 10.0 10.0 e) 9.0 f) 9.0 8.0 m/z 1721.9587 8.0 m/z 1721.9587 7.0 7.0 6.0 6.0 ppm 5.0 ror, 5.0 er 4.0 4.0 ss error, ppm Ma 3.0 Mass 3.0 2.0 2.0 1.0 1.0 0.0 0.0 1 10 100 1000 10000 1 10 100 1000 10000 Intensity ratio ratio Figure 3.